Research map / 3D vision / Depth and stereo
Depth Prediction without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videos
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add object motion modeling (ablation) | Monocular depth model · KITTI Eigen split, 80m | Abs Rel: 0.1417 -> 0.1412 | No clear effect |
| Add object motion modeling (ablation) | Monocular depth model · KITTI Eigen split, 80m | Sq Rel: 1.1385 -> 1.0258 | Helped |
| Add online refinement (ablation) | Monocular depth model · KITTI Eigen split, 80m | Abs Rel: 0.1417 -> 0.1231 | Helped |
| Combine object motion modeling with online refinement (ablation) | Monocular depth model · KITTI Eigen split, 80m | Abs Rel: 0.1417 -> 0.1087 | Helped |
| Add object motion modeling (ablation) | Cityscapes-trained depth model · KITTI evaluation | Abs Rel: 0.2054 -> 0.1876 | Helped |
| Add online refinement (ablation) | Cityscapes-trained depth model · KITTI evaluation | Abs Rel: 0.2054 -> 0.1696 | Helped |
| Combine object motion modeling with online refinement (ablation) | Cityscapes-trained depth model · KITTI evaluation | Abs Rel: 0.2054 -> 0.1529 | Helped |
| Use 3D object motion modeling | Monocular depth and ego-motion estimation · Dynamic scenes | depth prediction quality | Helped |
| Use online refinement during inference | Monocular depth model · Fetch Indoor Navigation, Cityscapes-trained | depth prediction quality | Helped |
| Apply online refinement while the camera is static | Monocular depth model · Static-camera inference | depth prediction quality | Hurt |
| Add a learned object-size constraint | 3D object motion model · Moving objects with ambiguous apparent motion | depth prediction quality | Helped |
| Train on Cityscapes and test on KITTI without fine-tuning | Ours (M+R) · Cityscapes-trained, KITTI evaluation | Abs Rel: 0.2054 -> 0.1529 | Helped |
Closest papers by meaning
- RM-Depth: Unsupervised Learning of Recurrent Monocular Depth in Dynamic Scenes (CVPR 2022 · similarity 0.91)
- Vision-Language Embodiment for Monocular Depth Estimation (CVPR 2025 · similarity 0.88)
- From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior (CVPR 2024 · similarity 0.88)
- Seurat: From Moving Points to Depth (CVPR 2025 · similarity 0.88)
- Mining Supervision for Dynamic Regions in Self-Supervised Monocular Depth Estimation (CVPR 2024 · similarity 0.88)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.